Type I Interferons Regulate the Magnitude and Functionality of Mouse Polyomavirus-Specific CD8 T Cells in a Virus Strain-Dependent Manner
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
crossmark
Type I Interferons Regulate the Magnitude and Functionality of
Mouse Polyomavirus-Specific CD8 T Cells in a Virus Strain-
Dependent Manner
Qingsong Qin,a Shwetank,a Elizabeth L. Frost,b Saumya Maru,a Aron E. Lukachera
Department of Microbiology and Immunology, The Pennsylvania State University College of Medicine, Hershey, Pennsylvania, USAa; Immunology and Molecular
Pathogenesis Graduate Program, Emory University, Atlanta, Georgia, USAb
ABSTRACT
Mouse polyomavirus (MPyV) is a ubiquitous persistent natural mouse pathogen. A glutamic acid (E)-to-glycine (G) difference at
Downloaded from http://jvi.asm.org/ on January 17, 2021 by guest
position 91 of the VP1 capsid protein shifts the profile of tumors induced by MPyV from an epithelial to a mesenchymal cell ori-
gin. Here we asked if this tropism difference affects the MPyV-specific CD8 T cell response, which controls MPyV infection and
tumorigenesis. Infection by the laboratory MPyV strain RA (VP1-91G) or a strain A2 mutant with an E-to-G substitution at VP1
residue 91 [A2(91G)] generated a markedly smaller virus-specific CD8 T cell response than that induced by A2(VP1-91E) infec-
tion. Mutant A2(91G)-infected mice showed a higher frequency of memory precursor (CD127hi KLRG1lo) CD8 T cells and a
higher recall response than those of A2-infected mice. Using T cell receptor (TCR)-transgenic CD8 T cells and immunization
with peptide-pulsed dendritic cells, we found that early bystander inflammation associated with A2 infection contributed to re-
cruitment of the larger MPyV-specific CD8 T cell response. Beta interferon (IFN-) transcripts were induced early during A2 or
A2(91G) infections. IFN- inhibited replication of A2 and A2(91G) in vitro. Using mice lacking IFN-␣ receptors (IFNARⴚ/ⴚ),
we showed that type I IFNs played a role in controlling MPyV replication in vivo but differentially affected the magnitude and
functionality of virus-specific CD8 T cells recruited by A2 and A2(91G) viral infections. These data indicate that type I IFNs are
involved in protection against MPyV infection and that their effect on the antiviral CD8 T cell response depends on capsid-medi-
ated tropism properties of the MPyV strain.
IMPORTANCE
Isolates of the human polyomavirus JC virus from patients with the frequently fatal demyelinating brain disease progressive
multifocal leukoencephalopathy (PML) carry single amino acid substitutions in the domain of the VP1 capsid protein that binds
the sialic acid moiety of glycoprotein/glycolipid receptors on host cells. These VP1 mutations may alter neural cell tropism or
enable escape from neutralizing antibodies. Changes in host cell tropism can affect recruitment of virus-specific CD8 T cells.
Using mouse polyomavirus, we demonstrate that a single amino acid difference in VP1 known to shift viral tropism profoundly
affects the quantity and quality of the anti-polyomavirus CD8 T cell response and its differentiation into memory cells. These
findings raise the possibility that CD8 T cell responses to infections by human polyomaviruses may be influenced by VP1 muta-
tions involving domains that engage host cell receptors.
B inding specificity among polyomaviruses is determined by in-
teraction of the VP1 major capsid protein with host cell gan-
gliosides having particular terminal sialic acid linkages (1). The
2,6)-linked sialic acids, thereby preventing binding of such
branched structures by LP strains; however, binding to ganglio-
sides with sialic acid (␣-2,3)-linked to galactose is retained for
gangliosides GD1a and GT1b are required for transport of mouse virion uptake into an infectious pathway (9, 10). Interestingly,
polyomavirus (MPyV) to the endoplasmic reticulum (2, 3). Dis- MPyVs isolated from feral mice have exclusively E-91 VP1s, an
crete amino acid differences in the receptor binding site of VP1 unexpected finding given that such LP viruses are potentially
impart important biological differences, including profound dif- more oncogenic than G-91 SP viruses (11).
ferences in pathogenicity (4). Replacement of the glycine (G) at The human polyomavirus JC virus (JCV) is a frequent member
position 91 of VP1 of the laboratory-derived small-plaque (SP) of the human virome (12). Mutations of JCV capsid protein VP1
MPyV strain RA with glutamic acid (E), the amino acid at this
position in the naturally occurring large-plaque (LP) strain PTA,
was sufficient to convert it into a strain with an LP morphology Received 31 January 2016 Accepted 10 March 2016
and to alter the profile of induced tumors from a mesenchymal to Accepted manuscript posted online 16 March 2016
an epithelial cell lineage (5). Alternatively, substitution of G for E Citation Qin Q, Shwetank, Frost EL, Maru S, Lukacher AE. 2016. Type I interferons
regulate the magnitude and functionality of mouse polyomavirus-specific CD8 T
at position 91 in VP1 in PTA had the opposite effect on plaque size cells in a virus strain-dependent manner. J Virol 90:5187–5199.
and tumorigenicity (6, 7). In SP strains, VP1 capsids with G-91 doi:10.1128/JVI.00199-16.
interact with branched (␣-2,6)-linked sialyloligosaccharides, Editor: G. McFadden
which may act as pseudoreceptors by binding cell surface glyco- Address correspondence to Aron E. Lukacher, alukacher@hmc.psu.edu.
proteins that divert virions into noninfectious pathways (8). An E Copyright © 2016, American Society for Microbiology. All Rights Reserved.
at this position in VP1 leads to electrostatic repulsion of the (␣-
May 2016 Volume 90 Number 10 Journal of Virology jvi.asm.org 5187Qin et al.
involving the sialic acid cell receptor binding domain are detected viral lysate at 4°C for 1 h, rinsed twice and cultured in Dulbecco’s modified
only in patients diagnosed with progressive multifocal leukoen- Eagle’s medium (DMEM) (Invitrogen Life Technologies) containing 5%
cephalopathy (PML), a frequently fatal demyelinating disease of fetal bovine serum (FBS), penicillin-streptomycin (100 U/ml; Mediat-
the central nervous system whose incidence is increasing in indi- ech), and 0.5% low-melting-temperature agarose for 5 to 6 days, fixed
viduals receiving immunomodulatory therapeutics for autoim- with 3.7% formaldehyde in phosphate-buffered saline (PBS), and stained
with 0.03% methylene blue in PBS, and plaques were then counted. For
mune diseases (13, 14). These VP1 mutations have been proposed
one-step growth curve assay of viruses, normal mouse mammary gland
to render JCV neurotropic but also appear to enable escape from
(NMuMG) epithelial cells (American Type Culture Collection) were in-
humoral immunity (13). Accumulating evidence supports the fected with MPyV [A2, RA, or A2(91G)] at a multiplicity of infection
likelihood that CD8 T cells are essential for controlling JCV infec- (MOI) of 1 and cultured in DMEM containing 10% FBS. Cells and super-
tion, preventing PML, and promoting recovery from PML (15, natant were harvested at the indicated days postinfection (p.i.), and after
16). Whether changes in tropism caused by VP1 mutations may freezing and thawing, the viral lysate was titrated by plaque assay. For
also affect anti-JCV CD8 T cell responses is not known. quantitative PCR (qPCR), DNA was extracted from organs or cell cultures
The fate and function of pathogen-specific CD8 T cells are by use of a DNA extraction kit (Promega), and viral genome copies were
affected by the duration of antigen availability, antigen levels, and quantified by qPCR as previously described (35).
Downloaded from http://jvi.asm.org/ on January 17, 2021 by guest
inflammatory factors (17–21). Memory CD8 T cells generated in Mice and infections. Female C57BL/6Ncr (B6) mice were purchased
response to infections that are cleared after acute infection self- from the National Cancer Institute (Frederick, MD). B6.129S2-
renew in an antigen-independent manner. We previously re- Ifnar1tm1Agt/Mmjax (IFNAR⫺/⫺) mice from The Jackson Laboratory
were backcrossed with C57BL/6 mice for 5 generations. Thy1.1 TCR-I-
ported that CD8 T cells recruited early in a high-dose MPyV in-
transgenic mice, bearing the T cell receptor (TCR) specific for LT206 –215
fection undergo exhaustion during the persistent phase of
of SV40, were described previously (36). Adult mice (8 to 12 weeks old)
infection (22); however, the less inflammatory environment asso- were inoculated in the hind footpads with 1.0 ⫻ 106 PFU of MPyV. For
ciated with early infection after low-dose MPyV inoculation fa- recall experiments, mice persistently infected (⬎30 days) with MPyV re-
vors differentiation of memory CD8 T cells with improved effec- ceived 1 ⫻ 106 PFU of rVSV-LT359 through hind footpad injection. All
tor function (23). Different pathogen-induced inflammatory protocols using mice were approved by the Institutional Animal Care and
environments play an important role in controlling proliferation Use Committee of The Pennsylvania State University College of Medicine.
and effector and memory differentiation of CD8 T cells (17, 24). Flow cytometric analysis. Red blood cell (RBC)-lysed blood and
In the case of lymphocytic choriomeningitis virus (LCMV), spleen samples were stained for 30 min at 4°C with monoclonal antibodies
changes in virus tropism caused by point mutations in the poly- (MAbs) against the following molecules: CD44 (clone IM7; BD), CD127
merase and glycoprotein capsid proteins resulted in dramatic dif- (clone A7R34; Biolegend), killer cell lectin-like receptor G1 (KLRG1)
ferences in viral loads and inflammatory environments and led to (clone 2F1; BD), and CD8␣ (clone 53-6.7; eBioscience). H-2Db LT359
and H-2Db LT206 tetramers (35) were provided by the NIH Tetramer
dysfunction and deletion of virus-specific CD8 T cells (i.e., T cell
Core Facility (Atlanta, GA). Samples were acquired on an LSRII instru-
exhaustion) (25–27). Type I interferons (IFN-I), interleukin-10
ment (BD), and data were analyzed using FlowJo software (TreeStar, Inc.,
(IL-10), and IL-12 are among the inflammatory factors that con- Ashland, OR). For intracellular cytokine staining (ICS), cells were stimu-
tribute to proliferation and differentiation of CD8 T cells during lated for 5 h with the LT359 –368 peptide (1 M) in the presence of
early infection (24, 28–30), and they also shape memory CD8 T Golgiplug (BD). Cells were surface stained, permeabilized, and fixed by
cell differentiation (31, 32). use of a Cytofix/Cytoperm kit (BD) and then stained with MAbs specific
In this study, we demonstrate that a single amino acid change for IFN-␥ (clone XMG1.2; eBioscience), IL-2 (clone JES6-5H4; BD), and
(E to G) at position 91 in VP1 alters the quantity and quality of the tumor necrosis factor alpha (TNF-␣) (clone TN3-19.12; BD). Fluores-
anti-MPyV CD8 T cell response. Moreover, we found that IFN-I is cence-minus-one (FMO) samples were used to set gates for each surface
induced by MPyV infection and regulates the magnitude and molecule examined (i.e., CD127, KLRG1, CD8, and CD44); for ICS, un-
function of virus-specific CD8 T cells but does so in a manner stimulated samples served as negative controls, and phorbol 12-myristate
dependent on this E-to-G variation in VP1. These findings have 13-acetate (PMA) (50 ng/ml)- and ionomycin (ION) (2 g/ml)-treated
samples served as positive controls. Nucleated cells from blood and spleen
important implications regarding the impact of VP1 mutations on
samples were counted with a TC20 automated cell counter (Bio-Rad). The
recruitment of protective CD8 T cells against human polyomavi- number of CD8 T cells in blood was determined by use of a flow cytometry
rus infections. absolute count standard (Bands Laboratories, Inc.).
Adoptive T cell transfer. CD8 T cells were purified from Thy1.1⫹
MATERIALS AND METHODS TCR-I-transgenic mice by use of a negative-selection CD8 T cell isolation
Viruses. Stocks of MPyV (A2 and RA) were prepared in baby mouse kit (Miltenyi Biotec) according to the manufacturer’s instructions. For all
kidney cells as described previously (33). Site-directed mutagenesis using cell transfer experiments, ⬎95% of donor TCR-I cells were CD44lo. One
previously published primers (7) was used to replace the codon for E with hundred TCR-I cells were transferred 1 day prior to infection. Recipient
that for G at position 91 of VP1 in the A2 genome; this mutant virus is mice were inoculated with 5 ⫻ 105 PFU of MPyV.LT206 or MPyV.LT206
designated A2(91G). The MPyV.LT206 mutant virus was created by re- (91G). TCR-I cells in 100 l of blood were quantified by flow cytometry at
placing the sequence encoding the Db-restricted LT359 –368 CD8 T cell day 8 p.i. For analysis of cell division, 1 ⫻ 106 TCR-I cells were labeled with
epitope from MPyV (SAVKNYCSKL) with that of the Db-restricted CD8 5 M carboxyfluorescein succinimidyl ester (CFSE) (Invitrogen) for 20
T cell epitope LT206 –215 from simian virus 40 (SV40) (SAINNYAQKL). min at room temperature and then injected intravenously (i.v.) into the
A2 and MPyV.LT206 have similar growth kinetics in vitro and in vivo (22). tail vein at day 4 p.i. The extent of CFSE dilution was determined by flow
A recombinant vesicular stomatitis virus carrying the LT359 –368 se- cytometry at day 6 p.i.
quence (rVSV-LT359) was generated as described previously (23). Dendritic cell immunization. Bone marrow-derived dendritic cells
Quantitation of MPyV by plaque assay and qPCR. Infectious virus (BMDCs) were cultured in vitro as previously described (37) and matured
was titrated by plaque assay on BALB/3T3 clone A31 cells as previously by exposure to 100 ng/ml lipopolysaccharide (LPS; Sigma-Aldrich) at day
described (34), with the following modifications. Subconfluent A31 cell 7 of culture. Cells were harvested the next day and assessed by flow cytom-
monolayers in 6-well tissue culture plates were inoculated with 200 l of etry. More than 90% of cells were F4/80lo MHC-IIhi CD11chi. BMDCs
5188 jvi.asm.org Journal of Virology May 2016 Volume 90 Number 10IFN-I Regulates the MPyV-Specific CD8 T Cell Response
TABLE 1 Primers used for quantitative RT-PCR
Target Forward primer Reverse Primer
IL-4 TCGGCATTTTGAACGAGGTC GAAAAGCCCGAAAGAGTCTC
IL-10 ACCTGGTAGAAGTGATGCCC GACACCTTGGTCTTGGAGCTT
IL-12 p35 AAATGAAGCTCTGCATCCTGC TCACCCTGTTGATGGTCACG
IL-12 p40 CCACTCACATCTGCTGCTCCACAA CAGTTCAATGGGCAGGGTCTCCTC
IFN- ACTCATGAAGTACAACAGCTACG GGCATCAACTGACAGGTCTT
TNF-␣ TACTGAACTTCGGGGTGATTGGTCC CAGCCTTGTCCCTTGAAGAGAACC
IFN-␥ TGAACGCTACACACTGCATCTTGG CGACTCCTTTTCCGCTTCCTGAG
18S rRNA CACTTTTGGGGCCTTCGTGT AGGCCCAGAGACTCATTTCTT
were pulsed with the LT206 peptide (10 M) at 37°C for 1.5 h and then (RA) showed differences in viral dissemination and in the magni-
washed, and 2.0 ⫻ 105 cells were injected i.v. into the tail vein in B6 mice. tude of the strongly dominant Db-restricted CD8 T cell response
Downloaded from http://jvi.asm.org/ on January 17, 2021 by guest
Three hours later, mice were injected with 1 ⫻ 106 PFU MPyV A2 or to amino acids 359 to 368 of the large T (LT) antigen, designated
A2(91G) in the hind footpads. The frequency of DbLT206 tetramer⫹ DbLT359 (48). To control for differences in the noncoding control
CD8⫹ cells in blood was analyzed by flow cytometry.
region and the few differences in coding regions between strains
Quantitation of cytokine mRNAs by qPCR. Total RNA was isolated
from spleens by use of a SimplyRNA kit (Promega) according to the man-
A2 and RA, we used site-directed mutagenesis to change the glu-
ufacturer’s instructions. One microgram of RNA was used to synthesize tamic acid (E) at position 91 of VP1 in A2 to glycine (G), as pre-
cDNA by use of a high-capacity cDNA reverse transcription kit (Thermo viously described (7). This A2(91G) mutant virus converted the
Scientific) according to the manufacturer’s instructions. Transcripts for parental A2 virus from LP to SP, confirming previous reports (6,
cytokine genes (IL-4, IL-10, IL-12, IFN-, TNF-␣, and IFN-␥) were quan- 7) that this single amino acid difference in VP1 is solely responsi-
tified by qPCR with SYBR green master mix (Roche), using an ABI Prism ble for the small-plaque morphology of the RA strain (Fig. 1A).
5700 sequence detection system (Applied Biosystems). The oligonucleo- The mutant A2(91G) exhibited in vitro growth kinetics similar to
tide primers used for qPCR are shown in Table 1. Primers for IL-4 (38), those of A2 and SP strain RA (Fig. 1B). However, viral loads in the
IL-12 p40 (39), IL-12 p35 (40), IFN-␥ (41), TNF-␣ (42), and 18S rRNA
spleens and kidneys of mice infected by the RA or A2(91G) virus
(43) were previously published. Primers for IFN- were validated by IDT.
Primers for IL-10 were designed by use of the NCBI Primer Blast tool and showed minimal differences over the course of infection (Fig. 1C)
then validated. Relative fold changes were determined using the threshold and were over 100-fold lower than those in mice infected by A2
cycle (2⫺⌬⌬CT) method, with the 18S rRNA gene used as a reference gene virus during the acute phase of infection. The frequency and mag-
for normalization, as previously described (44). Briefly, amounts of cyto- nitude of the DbLT359-specific CD8 T cell response in mice in-
kine mRNA were normalized to the amount of internal 18S rRNA. The fected by the RA or A2(91G) virus were significantly lower than
fold change of cytokine mRNA for each infected mouse was calculated by those in mice infected by the A2 virus, but notably, the kinetics of
dividing the amount of cytokine mRNA in infected mice by that in unin- T cell expansion and contraction in response to infection by each
fected mice. of these viruses were similar (Fig. 1D). To address the possibility
Immunofluorescence (IF) assay. Primary mouse embryonic fibro-
that less replication by A2(91G) may itself be responsible for the
blast (MEF) cells prepared from 13-day-old B6 mice were pretreated with
500 U/ml recombinant mouse IFN- purified from mammalian cell cul- smaller magnitude of the anti-MPyV CD8 T cell response, mice
ture (PBL Interferon Source, Piscataway, NJ) for 24 h. Cells were infected were inoculated with a 10-fold higher dose of A2(91G) than that of
with MPyV at an MOI of 0.01 and cultured with or without IFN- (500 the A2 strain. We found that the kinetics of viral replication and
U/ml). At day 3 p.i., cells were fixed at room temperature for 10 min with numbers of DbLT359-specific CD8 T cells remained significantly
2% paraformaldehyde in PBS, stained with the pan-MPyV T antigen- lower than those for A2-infected mice (data not shown). Given
specific F4 MAb (45) followed by goat anti-mouse IgG conjugated to that low-dose A2 virus inoculation yields a different DbLT359-
Alexa 594 (Invitrogen), and mounted with Prolong reagent with DAPI specific CD8 T cell response profile (delayed kinetics of expansion
(4=,6-diamidino-2-phenylindole) (Invitrogen). but an equal peak magnitude during acute infection) (23), these
Statistical analyses. Statistical significance was determined by un-
results indicated that the tropism difference imparted by the E-
paired, two-tailed Student’s t test, assuming equal variance. A P value
of ⬍0.05 was considered statistically significant. The 50% effective con- to-G switch at position 91 of VP1 negatively affected the size of the
centration of peptide (EC50) for the IFN-␥ stimulation assay was calcu- MPyV-specific CD8 T cell response.
lated according to nonlinear trajectory analysis by using GraphPad Prism MPyV-specific CD8 T cells recruited by A2(91G) virus infec-
software (San Diego, CA). tion exhibit superior functionality. During persistent infection
with LCMV strain cl13, a high viral load correlates with a loss of
RESULTS CD8 T cell polycytokine functionality and proliferative potential
E-to-G variation at position 91 of VP1 changes viral spread and (47). We next asked whether the 2-log difference in viral loads
the magnitude of the MPyV-specific CD8 T cell response. The during acute infection between the SP viruses RA and A2(91G)
magnitude of the CD8 T cell response correlates with the initial and the LP virus A2 affected the functional competence of the
viral load and duration of infection in various viral infection mod- MPyV-specific CD8 T cell response. During the acute (day 8 p.i.)
els (46, 47). LP strains of MPyV disseminate more extensively than and persistent (day 60 p.i.) phases of infection, A2(91G) virus-
do SP strains (5, 6). Administration of a low dose of the LP strain infected mice showed a lower frequency and number of splenic
A2 delays the kinetics of expansion, but not the magnitude, of the DbLT359-specific CD8 T cells than those for A2 virus-infected
MPyV-specific CD8 T cell response (23). We therefore asked mice, as detected by DbLT359 tetramers and ex vivo LT359 pep-
whether B6 mice infected with an LP strain (A2) or an SP strain tide-stimulated intracellular IFN-␥ and TNF-␣ production
May 2016 Volume 90 Number 10 Journal of Virology jvi.asm.org 5189Qin et al.
Downloaded from http://jvi.asm.org/ on January 17, 2021 by guest
FIG 1 E-to-G substitution at position 91 of VP1 limits viral spread and the magnitude of the MPyV-specific CD8 T cell response. (A) Plaque formation at day 6 p.i. by
A2, A2(91G), and RA strain viruses on A31 cells. (B) One-step growth curve assay of A2, RA, and A2(91G) in NMuMG cells infected at an MOI of 1. Viral titers were
measured by plaque assay at the indicated days p.i. (C) Viral DNAs in spleens and kidneys were measured by qPCR at the indicated days p.i. after A2, RA, or A2(91G)
inoculation (1.0 ⫻ 106 PFU) into hind footpads. (D) Frequencies (left) and numbers (right) of splenic DbLT359 tetramer⫹ CD8⫹ cells at the indicated days p.i. Data are
the means ⫾ standard errors of the means (SEM) for at least three mice per group and are representative of three experiments. *, P ⬍ 0.05.
(Fig. 2A). Comparison of the capacity of LT359-specific CD8 T effector cells (EECs), CD127lo KLRG1hi cells as short-lived ef-
cells to coproduce IFN-␥ and TNF-␣ (Fig. 2B, top and middle fector cells (SLECs), CD127hi KLRG1hi cells as double-positive
panels), however, revealed that MPyV-specific CD8 T cells re- effector cells (DPECs), and CD127hi KLRG1lo cells as memory
cruited by A2(91G) infection had an elevated capacity to co- precursor effector cells (MPECs). Using these markers, we ob-
produce these cytokines during persistent infection (Fig. 2B, served that acute infection with A2(91G) generated a signifi-
bottom panel). LT359 peptide titration further showed that cantly higher frequency of SLECs and a lower frequency of
MPyV-specific CD8 T cells recruited by A2(91G) infection dur- EECs than did A2 infection. In persistently infected mice, how-
ing persistent infection (day 60 p.i.) exhibited a higher antigen ever, A2(91G) recruited a higher frequency of MPECs and a
sensitivity (approximately 3-fold, as shown by EC50 values in lower frequency of SLECs than those with the A2 strain (Fig.
the legend to Fig. 2), a difference not seen for cells recruited 3A). The higher virus levels during acute A2 infection appear to
during acute infection (Fig. 2C). It merits highlighting that this favor differentiation of EECs over SLECs, whereas in persis-
difference in functional avidity by DbLT359-specific CD8 T tently infected mice, where levels of A2 and A2(91G) are similar
cells, albeit small, was not seen in mice given low- versus high- (Fig. 1C), A2 infection biases cells toward SLEC differentiation
dose A2 virus inocula (23). and A2(91G) infection tilts them toward MPEC differentia-
MPyV-specific CD8 T cells generated in response to A2 and tion.
A2(91G) infections also exhibited phenotypic differences indica- We next tested the prediction that this MPEC bias by MPyV-
tive of skewed memory T cell differentiation. Memory precursors specific CD8 T cells in mice persistently infected with A2(91G)
and terminally differentiated CD8 T cells can be distinguished was associated with an augmented recall capability upon antigen
based on their expression of CD127 (IL-7R␣) (49, 50) and killer challenge. A2- and A2(91G)-infected mice at day 84 p.i. were thus
cell lectin-like receptor G1 (KLRG1) (51). Phenotypic dissection infected with a recombinant VSV strain encoding the LT359
using these markers demarcates CD127lo KLRG1lo cells as early epitope (22). As shown in Fig. 3B, at 5 days postchallenge,
5190 jvi.asm.org Journal of Virology May 2016 Volume 90 Number 10IFN-I Regulates the MPyV-Specific CD8 T Cell Response
Downloaded from http://jvi.asm.org/ on January 17, 2021 by guest
FIG 2 A2(91G) infection recruits memory CD8 T cells having higher polycytokine functionality and functional avidity. (A) Representative dot plots of splenic
DbLT359 tetramer⫹ CD8 T cells surface stained with anti-CD44 and of intracellular costaining with anti-IFN-␥ and anti-TNF-␣ of anti-CD8 surface-stained
spleen cells after LT359 peptide stimulation at days 8 and 60 p.i. (no IFN-␥⫹ or TNF-␣⫹ cells were detected in the absence of LT359 peptide [data not shown]).
(B) Numbers of DbLT359 tetramer⫹, IFN-␥⫹, and IFN-␥⫹ TNF-␣⫹ CD8 T cells at days 8 and 60 p.i. (top and middle) and ratios of TNF-␣⫹ to IFN-␥⫹ CD8 T
cells (bottom). Values represent the means and SEM for three experiments using 3 or 4 mice per group, *, P ⬍ 0.05; **, P ⬍ 0.005; NS, not significant. (C) LT359
peptide dose-response curves for intracellular IFN-␥ staining of splenic CD8 T cells from A2- and A2(91G)-infected mice at day 8 p.i. (top) and day 60 p.i.
(bottom). EC50 values were as follows: at day 8 p.i., 9.0 ⫻ 10⫺10 M for A2 and 8.5 ⫻ 10⫺10 M for A2(91G) (P ⫽ 0.96); and at day 60 p.i., 8.71 ⫻ 10⫺10 M for A2
and 3.3 ⫻ 10⫺10 M for A2(91G) (P ⫽ 0.017). Experiments were repeated twice, with 3 mice per group.
DbLT359-specific CD8 T cells were expanded to a significantly 36). We designated this virus LT206(91G). MPyV.LT206 and
greater extent in A2(91G)-infected mice than in A2-infected LT206(91G) have replication kinetics similar to those of A2 and
mice. Taken together, the data show that E-to-G variation at A2(91G), respectively (data not shown). Adoptive transfer of
position 91 of VP1 is associated with superior functionality of Thy1.1⫹ TCR-I cells into B6 (Thy1.2) mice prior to infection re-
the MPyV-specific CD8 T cell response during persistent infec- capitulated the observation made in Fig. 1D that peak accumula-
tion. tion of virus-specific CD8 T cells during acute infection was higher
Inflammatory environments created by A2 versus A2(91G) for MPyV.LT206(E-91) than for LT206(91G) (Fig. 4A). Using
infection differentially affect the MPyV-specific CD8 T cell re- CFSE-labeled donor TCR-I cells, we determined that this higher-
sponse. Early inflammation during viral infection plays a role in magnitude expansion could be accounted for by more efficient
regulating CD8 T cell proliferation (17). Thus, we asked whether expansion of anti-MPyV CD8 T cells in mice infected by the E-91
the inflammatory environment associated with A2 infection and VP1 than the G-91 VP1 virus (Fig. 4B).
A2(91G) infection affected the proliferation of naive virus-specific To determine whether the A2 infection-associated inflamma-
CD8 T cells. We introduced the G-91 VP1 mutation into the pre- tory environment also contributed to more efficient recruitment
viously characterized MPyV.LT206 virus, an A2 mutant in which of naive virus-specific CD8 T cells, we immunized mice with
the LT359 peptide sequence was replaced with that of the SV40 LT206 peptide-pulsed BMDCs in the setting of infection by A2 or
LT206 epitope recognized by TCR-I-transgenic CD8 T cells (22, A2(91G) virus, both of which lack the DbLT206 epitope. No cross-
May 2016 Volume 90 Number 10 Journal of Virology jvi.asm.org 5191Qin et al.
Downloaded from http://jvi.asm.org/ on January 17, 2021 by guest
FIG 3 A2(91G) infection-recruited CD8 T cells exhibit a skewed CD127hi KLRG1lo phenotype and have a larger recall response. (A) At days 8 and 84 p.i., spleen
cells were surface stained with DbLT359 tetramers and MAbs to KLRG1 and CD127. (Left) Contour plots representative of 3 to 5 mice per time point and three
independent experiments; (right) frequencies of LT359-specific CD8 T cells with the MPEC (KLRG1lo CD127hi), DPEC (KLRG1hi CD127hi), SLEC (KLRG1hi
CD127lo), and EEC (KLRG1lo CD127lo) phenotypes. Values indicate means and SEM. *, P ⬍ 0.05; **, P ⬍ 0.005. (B) At 80 days p.i. with A2 or A2(91G), B6 mice
were challenged with rVSV-LT359. (Left) Representative mean frequencies of DbLT359 tetramer⫹ CD44⫹ CD8 T cells in blood before and 5 days after challenge;
(right) pre- versus postchallenge fold expansion, with each dot corresponding to an individual mouse and horizontal lines indicating means ⫾ SEM. Data were
collected from two independent experiments (n ⫽ 4 or 5 mice for each group).
reactivity is seen by CD8 T cells specific for the DbLT206 and nificant difference was observed in induction of IFN- transcripts
DbLT359 epitopes (22; data not shown). This approach allowed us between A2 and A2(91G) virus-infected mice (Fig. 6A). Interest-
to eliminate differences in cognate viral antigen levels resulting ingly, by day 6 p.i., mice infected with A2 virus showed a dramatic
from A2 versus A2(91G) infection. As shown in Fig. 5, the inflam- upregulation of transcripts for IL-12 p40, the subunit shared by
matory environment associated with A2 infection drove a signif- IL-12 and IL-23, as well as transcripts for IL-4 and IL-10; strong
icantly larger DbLT206 CD8 T cell expansion than that with induction of transcripts for these cytokines was not seen in
A2(91G) infection by day 6 p.i. These results raised the possibility A2(91G)-infected mice (Fig. 6A). This difference in profile for
that A2 and A2(91G) elicited different types or amounts of cyto- these cytokines between A2 and A2(91G) during acute infection
kines early in infection that affected the expansion of MPyV-spe- may influence the effector versus memory differentiation of
cific CD8 T cells. MPyV-specific CD8 T cells. To determine whether type I interfer-
Type I IFN controls MPyV replication in vitro and in vivo. ons inhibited MPyV replication, primary mouse embryonic fibro-
Using quantitative PCR, we assessed the mRNA levels for candi- blasts (MEFs) were infected with either A2 or mutant A2(91G)
date cytokines (IL-4, IL-10, IL-12, IFN-, TNF-␣, and IFN-␥) virus in the presence or absence of mouse IFN-. IFN- inhibited
potentially elicited early during MPyV infection and compared infection by both the A2 and A2(91G) viruses. Approximately
the levels of these transcripts in MPyV-infected and uninfected 15% of MEFs among untreated cells were T antigen-positive, while
mice. Of the cytokine transcripts examined at days 3 and 6 p.i., fewer than 1% of cells among IFN--treated cells were positive
IFN- mRNA levels were elevated in infected mice at both time (Fig. 6B). The antiviral effect of IFN- was confirmed by enumer-
points, although they were markedly higher at day 3 p.i.; no sig- ation of MPyV genome copies by qPCR (Fig. 6C). IFN--medi-
5192 jvi.asm.org Journal of Virology May 2016 Volume 90 Number 10IFN-I Regulates the MPyV-Specific CD8 T Cell Response
Downloaded from http://jvi.asm.org/ on January 17, 2021 by guest
FIG 4 Naive MPyV-specific CD8 T cells are recruited more efficiently by infection with A2(91E) virus than by that with A2(91G) virus. (A) Experimental
setup. One hundred naive Thy1.1 TCR-I cells were injected i.v. into the tail vein in B6 (Thy1.2) mice 1 day before inoculation with 5 ⫻ 105 PFU of
MPyV.LT206(91E) or MPyV.LT206(91G). The number of TCR-I cells in blood was determined by flow cytometry at day 8 p.i. Data were collected from
three experiments with three mice per group. Each dot represents an individual mouse, with horizontal lines indicating means ⫾ SEM. ***, P ⬍ 0.0005.
(B) (Top) Experimental setup. A total of 1 ⫻ 106 naive TCR-I cells labeled with CFSE were transferred i.v. into B6 mice inoculated 5 days earlier with
MPyV.LT206(91E) or MPyV.LT206(91G). (Bottom left) After 40 h, the number of cell divisions of donor TCR-I cells was determined based on CFSE
dilution 2 days after cell transfer. (Bottom right) Mean numbers of cell divisions of TCR-I cells (and SEM) for three mice per group from two independent
experiments. **, P ⬍ 0.005.
ated antiviral effects were also observed in vivo. In spleens of B6 receptors (Fig. 7C and as shown by EC50 values in the legend to
versus IFNAR⫺/⫺ mice infected with A2 or A2(91G), viral levels in Fig. 7), as well as skewed differentiation toward the MPEC
IFNAR⫺/⫺ mice were approximately 1 log higher than those in B6 phenotype (Fig. 7D). Thus, MPyV infections of wild-type and
mice at day 4 p.i. (Fig. 6D), a time point preceding detection of IFNAR⫺/⫺ mice, irrespective of the 91E-to-G VP1 tropism dif-
MPyV-specific CD8 T cells (52). These results indicate that IFN-I ference, inversely affected the effector function and functional
controls viral replication early in the course of MPyV infection avidity of the virus-specific CD8 T cells.
and that it does so equivalently for infections with A2 and
A2(91G). DISCUSSION
Next, we compared MPyV-specific CD8 T cell responses in B6 In this study, we compared the CD8 T cell responses to MPyV
versus IFNAR⫺/⫺ mice infected by the A2 or A2(91G) virus. As strains with different receptor binding specificities and affinities.
shown in Fig. 7A and B, the absence of IFN-I receptors had oppo- The single-amino-acid E-to-G exchange at position 91 in VP1
site effects on the anti-MPyV CD8 T cell responses recruited in causes MPyV virions to bind not only to terminally unbranched
response to A2 and A2(91G) infections. Although the size of the sialic acid true cellular receptors but also to branched-chain sialic
DbLT359-specific CD8 T cell response was lower in A2-infected acid “pseudoreceptors,” resulting in attenuation of viral spread
IFNAR⫺/⫺ mice than B6 mice, the reverse was seen for A2(91G) and tumorigenicity (6, 9). We validated that the E-to-G substitu-
infection; this difference was seen in both acutely and persistently tion at VP1 position 91 dampened viral replication in vivo and that
infected mice (Fig. 7A and B). IFN-I receptor deficiency was it did so in adult B6 mice, which are highly resistant to MPyV
generally associated with an augmented cytokine-producing tumorigenesis (53). Here we found that this VP1 difference lim-
capability, which was particularly evident for polycytokine ited expansion of MPyV-specific CD8 T cells and endowed
production by CD8 T cells during persistent infection (Fig. them with superior function (more polycytokine production,
7B). Interestingly, memory CD8 T cells in mice persistently higher functional avidity, and a larger recall response) during
infected with either A2 or A2(91G) virus exhibited a shift to- persistent infection. We further found that IFN-I was induced
ward decreased viral antigen sensitivity in the absence of IFN-I early during MPyV infection and differentially regulated the
May 2016 Volume 90 Number 10 Journal of Virology jvi.asm.org 5193Qin et al.
Downloaded from http://jvi.asm.org/ on January 17, 2021 by guest
FIG 5 Inflammation associated with A2 and A2(91G) infections differentially promotes CD8 T cell expansion. (Left) Experimental setup. A total of 2 ⫻ 105
BMDCs pulsed with LT206 peptide were injected i.v. into the tail vein in B6 mice. Three hours later, mice were inoculated with 1.0 ⫻ 106 PFU of A2 or A2(91G)
in the hind footpads. (Right) The frequency of DbLT206 tetramer⫹ CD8 T cells in blood was analyzed at the indicated days p.i. Data were combined from three
independent experiments with three mice per group. Values indicate means and SEM. **, P ⬍ 0.005; NS, not significant.
magnitude and functionality of virus-specific CD8 T cells in a markedly upregulated during early acute infection by both
manner dependent on the tropism of viruses with an E versus G A2(91E) and A2(91G) viruses, whereas mRNAs for IL-12
at VP1 position 91. (and/or IL-23) were increased later in acute infection only by
CD8 T cell differentiation is programmed by three major A2(91E) virus.
factors: the strength and duration of peptide major histocom- Type I IFNs are comprised of a family of highly related mole-
patibility complex (MHC)-TCR engagement, the availability cules that program a state of resistance to intracellular pathogens
and constellation of costimulatory molecules from antigen- and serve to alarm cells of both innate and adaptive immunity of
presenting cells, and the profile of inflammatory cytokines. Ac- the presence of infections (57). In our study, we showed that
cumulating literature indicates that the pathogen-induced in- MPyV replication was directly inhibited by IFN- in vitro and that
flammatory environment plays a central role in guiding the viral levels were higher in IFNAR⫺/⫺ mice than in wild-type mice.
pathway for memory CD8 T cell differentiation (17, 24). Our IFN-I has been shown to promote IFN-␥ production by CD4 T
previous study showed that early virus-associated bystander cells through STAT4 activation (58), to enhance the terminal dif-
events induced by a low-dose inoculum of MPyV A2 generated ferentiation of dendritic cells (59), and to augment the prolifera-
virus-specific CD8 T cells with a superior capacity to produce tion of CD8 T cells (55). We showed that the tropism change due
IFN-␥. CD8 T cells primed by low-dose MPyV also expressed to the E-to-G difference at position 91 of VP1, while associated
higher levels of CD127 and the antiapoptotic protein Bcl-2, with a lower-magnitude CD8 T cell response, was equally capable
which are required for memory CD8 T cell survival during of eliciting IFN-. This discrepancy might be explained by the
persistent infection (23). The low-dose A2 strain inoculation timing of exposure of naive CD8 T cells to IFN-I. IFN-I may mit-
delayed the kinetics but did not change the magnitude of the igate CD8 T cell proliferation when T cells encounter IFN-I before
CD8 T cell response compared to high-dose inoculation of the engaging their cognate antigen (60). Because A2(91G) dissemi-
A2 strain virus (23). In sharp contrast, infection with the VP1 nates less efficiently than A2, naive anti-MPyV CD8 T cells re-
mutant A2(91G) virus, which spread poorly compared to the cruited in response to infection by A2(91G) may bind IFN-I be-
parental A2(91E) virus, generated an MPyV-specific CD8 T cell fore TCR activation. Thus, it is conceivable that polyomavirus
response with the same kinetics as that of the A2(91E) virus, VP1 variants that differ in type and/or affinity for host cell recep-
albeit of lower magnitude. Thus, the lower virus infection levels tor glycans may alter the fate of the antiviral CD8 T cell response
and lesser spread by A2(91G) were of themselves insufficient by changing when naive T cells engage IFN-I relative to antigen.
factors to account for the less efficient recruitment of anti- This possibility is supported by evidence that innate inflammatory
MPyV CD8 T cells. Using an adoptive T cell transfer approach signals induced by different pathogens differentially dictate the
and dendritic cell immunization, we demonstrated that the IFN-I dependence of CD8 T cell expansion and memory differen-
early inflammatory environment caused by a change in viral tiation (24). An increase in IL-12 (and/or IL-23) during acute
receptor binding contributed to a different magnitude and infection by A2(91E) virus may also have promoted MPyV-spe-
functionality of virus-specific CD8 T cells. IFN-I and IL-12 cific CD8 T cell expansion and effector differentiation.
have been described as a “third signal” to promote full activa- IFN-I plays an important role in the effector and memory fate
tion and development of CD8 T cells (54–56). Through cyto- decision process via regulating the secretion of particular cyto-
kine profiling, we found that the transcripts of IFN- were kines (e.g., IL-15) (61–63) and the expression of the transcription
5194 jvi.asm.org Journal of Virology May 2016 Volume 90 Number 10IFN-I Regulates the MPyV-Specific CD8 T Cell Response
Downloaded from http://jvi.asm.org/ on January 17, 2021 by guest
FIG 6 Type I interferons inhibit viral replication in vitro and in vivo. (A) Fold changes of the indicated cytokine mRNAs from spleens of infected mice at days
3 and 6 p.i. normalized to transcripts from spleens of uninfected mice. Data were collected from three independent experiments with three mice per group. Values
indicate means and SEM. *, P ⬍ 0.05. (B) MEFs were infected with A2 or A2(91G) virus at an MOI of 0.01, with and without addition of IFN-. MEFs were stained
with a pan-T-antigen MAb and counterstained with DAPI at 24 h p.i. Representative percentages of T antigen⫹ cells among DAPI⫹ cells (⬎250 cells counted)
from two experiments are shown. (C) Viral genome copies in the cell lysates used for panel B were quantified by qPCR at the indicated time points. Values indicate
means and SEM for triplicates in one experiment. This experiment was repeated twice. (D) Viral loads in spleens of infected B6 and IFNAR⫺/⫺ mice were
measured by qPCR at days 4 and 7 p.i. Values indicate means and SEM for 4 or 5 mice from each group. *, P ⬍ 0.05.
factors T-bet (51) and eomesodermin (64). IFN-I receptor signal- may also impair memory CD8 T cell functionality, as recently
ing regulates the expression of CD127 and KLRG1, which are mol- reported for chronic LCMV infection, where it has been proposed
ecules used to demarcate subsets of effector and memory T cells to limit immunopathology (65–67). IFN-I receptor signaling by
(49, 51). Our data showed that the E-to-G change at position 91 of CD8 T cells during acute LCMV infection upregulated T-bet and
VP1 shifted CD8 T cell differentiation toward SLECs during acute directed differentiation toward SLECs, while a subset of CD8 T
infection but toward MPECs during persistent infection. In con- cells exhibited the MPEC phenotype during chronic LCMV infec-
trast, IFN-I signaling skewed virus-specific CD8 T cell differenti- tion in both IFN-I receptor-sufficient and -deficient mice (68).
ation toward SLECs during persistent infection by A2 virus. The While a mechanistic understanding of how IFN-I modulates CD8
increased polycytokine functionality by MPyV-specific memory T cell differentiation remains unknown, our results reinforce the
CD8 T cells in mice lacking IFN-I receptors may be related to their concept that the impact of IFN-I on memory differentiation is
lower peak expansion and effector differentiation during acute pathogen dependent.
infection. IFN-I receptor signaling during persistent infection All MPyV isolates from feral mice have a glutamic acid at VP1
May 2016 Volume 90 Number 10 Journal of Virology jvi.asm.org 5195Qin et al.
Downloaded from http://jvi.asm.org/ on January 17, 2021 by guest
FIG 7 Type I interferons regulate CD8 T cell proliferation, functionality, and memory differentiation in a virus strain-dependent manner. B6 mice and
IFNAR⫺/⫺ mice were infected with 1.0 ⫻ 106 PFU of A2 or A2(91G) virus in the hind footpads. (A) The frequency of DbLT359 tetramer⫹ CD8 T cells in blood
was measured by flow cytometry at days 7 and 50 p.i. Data were collected from two independent experiments (3 to 5 mice/group). (B) Splenocytes from mice
inoculated 55 days earlier with A2 or A2(91G) were surface stained with anti-CD44, anti-CD8␣, and DbLT359 tetramers or stained for surface CD8␣ and
intracellular IFN-␥ and TNF-␣ after 5 h of stimulation with LT359 peptide or without peptide (data not shown; used to set gates). (Left) Numbers of splenic
DbLT359 tetramer⫹ CD8 T cells and LT359 peptide-stimulated IFN-␥⫹ and TNF-␣⫹ IFN-␥⫹ CD8 T cells. (Right) Ratios of TNF-␣⫹ to IFN-␥⫹ LT359 CD8 T
cells. (C) LT359 peptide dose-response curves for intracellular IFN-␥ staining of splenic CD8 T cells at day 55 p.i. The EC50 for A2-infected B6 mice was 8.87 ⫻
10⫺10 M, and that for A2-infected IFNAR⫺/⫺ mice was 2.25 ⫻ 10⫺9 M (P ⫽ 0.04). The EC50 for A2(91G)-infected B6 mice was 2.3 ⫻ 10⫺10 M, and that for
A2(91G)-infected IFNAR⫺/⫺ mice was 7.63 ⫻ 10⫺10 M (P ⫽ 0.0022). (D) At day 50 p.i., splenic DbLT359 tetramer⫹ CD8 T cells were stained for KLRG1 and
CD127; designations based on these molecules are described in the legend to Fig. 3A. Data are representative of three experiments using 3 or 4 mice/group. Values
indicate means and SEM. *, P ⬍ 0.05; **, P ⬍ 0.005; NS, not significant.
position 91 (11). From the perspective of virus-host coevolution, mutations in VP1 may serve to enable escape from neutralizing
it is seemingly counterintuitive that mice in the wild harbor MPyV JCV antibodies (71, 72). PML nonsurvivors had selectively im-
strains with a VP1 amino acid shared by highly tumorigenic lab- paired JCV-specific CD8 T cell responses (16, 73). We recently
oratory strains. Tumor induction by MPyV, however, is relegated reported that brain-resident MPyV-specific memory CD8 T
to neonatal infection of virus-free colonies of mice of tumor-sus- cells expressed higher-affinity TCRs than those of splenic CD8
ceptible strains or infection of immunodeficient naive adult mice T cells (74). Whether IFN-I is induced by JCV infection and
(69). In natural mouse populations, MPyV establishes a silent per- affects the functional integrity of CD8 T cells in PML patients
sistent infection. It has been proposed that differences in regula- in response to VP1 variant viruses is an open question. IFN-I
tory sequences among MPyV isolates from feral mice might offset receptor blockade has been reported to reverse CD8 T cell ex-
the viral oncogenicity by changing host range properties of the haustion and to control chronic LCMV infection (67, 75). Our
virus. The results presented here give the additional explanation study suggests that the IFN-I pathway may similarly warrant
that the relative impairment in virus-specific CD8 T cell immunity consideration as a therapeutic target for persistent polyomavi-
associated with infection by E91-VP1 strains in feral mice favors rus infection.
higher viral loads and intermouse transmission.
CD8 T cells play an essential role in controlling persistent ACKNOWLEDGMENTS
infection by polyomavirus (34). The high frequency of amino We thank Ge Jin for excellent technical support and animal handling and
acid mutations in VP1 close to or at the receptor binding site in Taryn Mockus for a critical review of the manuscript. We thank Ziaur
isolates of JCV from PML patients suggests that subtle changes Rahman (Department of Microbiology and Immunology, Penn State Col-
in binding affinity for sialylated receptors can significantly af- lege of Medicine) for generously providing IFNAR⫺/⫺ mice. We acknowl-
fect viral pathogenicity (70). Recent evidence indicates that edge the technical support of Nate Sheaffer, Jade Vogel, and Joe Bednar-
5196 jvi.asm.org Journal of Virology May 2016 Volume 90 Number 10IFN-I Regulates the MPyV-Specific CD8 T Cell Response
cyk of the Flow Cytometry Core Facility of The Penn State College of Berger JR, Ngo L, Koralnik IJ. 2011. Role of CD4⫹ and CD8⫹ T-cell
Medicine (Hershey, PA). responses against JC virus in the outcome of patients with progressive
multifocal leukoencephalopathy (PML) and PML with immune reconsti-
tution inflammatory syndrome. J Virol 85:7256 –7263. http://dx.doi.org
FUNDING INFORMATION /10.1128/JVI.02506-10.
This work, including the efforts of Aron Lukacher, was funded by HHS | 17. Obar JJ, Jellison ER, Sheridan BS, Blair DA, Pham QM, Zickovich JM,
NIH | National Institute of Allergy and Infectious Diseases (NIAID) Lefrancois L. 2011. Pathogen-induced inflammatory environment con-
(AI102543). This work, including the efforts of Aron Lukacher, was trols effector and memory CD8⫹ T cell differentiation. J Immunol 187:
funded by HHS | NIH | National Institute of Neurological Disorders and 4967– 4978. http://dx.doi.org/10.4049/jimmunol.1102335.
Stroke (NINDS) (NS088367). This work, including the efforts of Eliza- 18. Blair DA, Turner DL, Bose TO, Pham QM, Bouchard KR, Williams KJ,
beth Frost, was funded by HHS | NIH | National Institute of Neurological McAleer JP, Cauley LS, Vella AT, Lefrancois L. 2011. Duration of
Disorders and Stroke (NINDS) (NS083336). antigen availability influences the expansion and memory differentiation
of T cells. J Immunol 187:2310 –2321. http://dx.doi.org/10.4049
/jimmunol.1100363.
REFERENCES 19. Urban SL, Welsh RM. 2014. Out-of-sequence signal 3 as a mechanism for
1. Neu U, Stehle T, Atwood WJ. 2009. The Polyomaviridae: contributions of virus-induced immune suppression of CD8 T cell responses. PLoS Pathog
virus structure to our understanding of virus receptors and infectious entry. 10:e1004357. http://dx.doi.org/10.1371/journal.ppat.1004357.
Downloaded from http://jvi.asm.org/ on January 17, 2021 by guest
Virology 384:389 –399. http://dx.doi.org/10.1016/j.virol.2008.12.021. 20. Mueller SN, Ahmed R. 2009. High antigen levels are the cause of T cell
2. Tsai B, Gilbert JM, Stehle T, Lencer W, Benjamin TL, Rapoport TA. exhaustion during chronic viral infection. Proc Natl Acad Sci U S A 106:
2003. Gangliosides are receptors for murine polyoma virus and SV40. 8623– 8628. http://dx.doi.org/10.1073/pnas.0809818106.
EMBO J 22:4346 – 4355. http://dx.doi.org/10.1093/emboj/cdg439. 21. Wherry EJ, Ahmed R. 2004. Memory CD8 T-cell differentiation during
3. You J, O’Hara SD, Velupillai P, Castle S, Levery S, Garcea RL, Benja- viral infection. J Virol 78:5535–5545. http://dx.doi.org/10.1128/JVI.78.11
min T. 2015. Ganglioside and non-ganglioside mediated host responses to .5535-5545.2004.
the mouse polyomavirus. PLoS Pathog 11:e1005175. http://dx.doi.org/10 22. Wilson JJ, Pack CD, Lin E, Frost EL, Albrecht JA, Hadley A, Hofstetter
.1371/journal.ppat.1005175. AR, Tevethia SS, Schell TD, Lukacher AE. 2012. CD8 T cells recruited
4. Bauer PH, Bronson RT, Fung SC, Freund R, Stehle T, Harrison SC, early in mouse polyomavirus infection undergo exhaustion. J Immunol
Benjamin TL. 1995. Genetic and structural analysis of a virulence deter- 188:4340 – 4348. http://dx.doi.org/10.4049/jimmunol.1103727.
minant in polyomavirus VP1. J Virol 69:7925–7931. 23. Andrews NP, Pack CD, Vezys V, Barber GN, Lukacher AE. 2007. Early
5. Dubensky TW, Freund R, Dawe CJ, Benjamin TL. 1991. Polyomavirus virus-associated bystander events affect the fitness of the CD8 T cell re-
replication in mice: influences of VP1 type and route of inoculation. J sponse to persistent virus infection. J Immunol 178:7267–7275. http://dx
Virol 65:342–349. .doi.org/10.4049/jimmunol.178.11.7267.
6. Bauer PH, Cui C, Liu WR, Stehle T, Harrison SC, DeCaprio JA, 24. Thompson LJ, Kolumam GA, Thomas S, Murali-Krishna K. 2006.
Benjamin TL. 1999. Discrimination between sialic acid-containing recep- Innate inflammatory signals induced by various pathogens differentially
tors and pseudoreceptors regulates polyomavirus spread in the mouse. J dictate the IFN-I dependence of CD8 T cells for clonal expansion and
Virol 73:5826 –5832. memory formation. J Immunol 177:1746 –1754. http://dx.doi.org/10
7. Freund R, Garcea RL, Sahli R, Benjamin TL. 1991. A single-amino-acid .4049/jimmunol.177.3.1746.
substitution in polyomavirus VP1 correlates with plaque size and hemag- 25. Ahmed R, Hahn CS, Somasundaram T, Villarete L, Matloubian M,
glutination behavior. J Virol 65:350 –355. Strauss JH. 1991. Molecular basis of organ-specific selection of viral vari-
8. Qian M, Tsai B. 2010. Lipids and proteins act in opposing manners to ants during chronic infection. J Virol 65:4242– 4247.
regulate polyomavirus infection. J Virol 84:9840 –9852. http://dx.doi.org 26. Matloubian M, Kolhekar SR, Somasundaram T, Ahmed R. 1993. Mo-
/10.1128/JVI.01093-10. lecular determinants of macrophage tropism and viral persistence: impor-
9. Buch MH, Liaci AM, O’Hara SD, Garcea RL, Neu U, Stehle T. 2015. tance of single amino acid changes in the polymerase and glycoprotein of
Structural and functional analysis of murine polyomavirus capsid pro- lymphocytic choriomeningitis virus. J Virol 67:7340 –7349.
teins establish the determinants of ligand recognition and pathogenicity. 27. Wherry EJ, Blattman JN, Murali-Krishna K, van der Most R, Ahmed R.
PLoS Pathog 11:e1005104. http://dx.doi.org/10.1371/journal.ppat 2003. Viral persistence alters CD8 T-cell immunodominance and tissue
.1005104. distribution and results in distinct stages of functional impairment. J Virol
10. Stehle T, Harrison SC. 1996. Crystal structures of murine polyomavirus 77:4911– 4927. http://dx.doi.org/10.1128/JVI.77.8.4911-4927.2003.
in complex with straight-chain and branched-chain sialyloligosaccharide 28. Baca Jones C, Filippi C, Sachithanantham S, Rodriguez-Calvo T, Eh-
receptor fragments. Structure 4:183–194. http://dx.doi.org/10.1016 rhardt K, von Herrath M. 2014. Direct infection of dendritic cells during
/S0969-2126(96)00021-4. chronic viral infection suppresses antiviral T cell proliferation and induces
11. Carroll J, Dey D, Kreisman L, Velupillai P, Dahl J, Telford S, Bronson IL-10 expression in CD4 T cells. PLoS One 9:e90855. http://dx.doi.org/10
R, Benjamin T. 2007. Receptor-binding and oncogenic properties of .1371/journal.pone.0090855.
polyoma viruses isolated from feral mice. PLoS Pathog 3:e179. http://dx 29. Brooks DG, Trifilo MJ, Edelmann KH, Teyton L, McGavern DB,
.doi.org/10.1371/journal.ppat.0030179. Oldstone MB. 2006. Interleukin-10 determines viral clearance or persis-
12. Virgin HW. 2014. The virome in mammalian physiology and disease. Cell tence in vivo. Nat Med 12:1301–1309. http://dx.doi.org/10.1038/nm1492.
157:142–150. http://dx.doi.org/10.1016/j.cell.2014.02.032. 30. Xiao Z, Casey KA, Jameson SC, Curtsinger JM, Mescher MF. 2009.
13. Gorelik L, Reid C, Testa M, Brickelmaier M, Bossolasco S, Pazzi A, Programming for CD8 T cell memory development requires IL-12 or type
Bestetti A, Carmillo P, Wilson E, McAuliffe M, Tonkin C, Carulli JP, I IFN. J Immunol 182:2786 –2794. http://dx.doi.org/10.4049/jimmunol
Lugovskoy A, Lazzarin A, Sunyaev S, Simon K, Cinque P. 2011. Pro- .0803484.
gressive multifocal leukoencephalopathy (PML) development is associ- 31. Keppler SJ, Rosenits K, Koegl T, Vucikuja S, Aichele P. 2012. Signal 3
ated with mutations in JC virus capsid protein VP1 that change its receptor cytokines as modulators of primary immune responses during infections:
specificity. J Infect Dis 204:103–114. http://dx.doi.org/10.1093/infdis the interplay of type I IFN and IL-12 in CD8 T cell responses. PLoS One
/jir198. 7:e40865. http://dx.doi.org/10.1371/journal.pone.0040865.
14. Bayliss J, Harrison E, McLean CA. 2011. Progressive multifocal leuko- 32. Way SS, Havenar-Daughton C, Kolumam GA, Orgun NN, Murali-
encephalopathy development is associated with mutations in JC virus cap- Krishna K. 2007. IL-12 and type-I IFN synergize for IFN-␥ production by
sid protein VP1 that change the receptor specificity of the virus. J Infect Dis CD4 T cells, whereas neither are required for IFN-␥ production by CD8 T
204:1643–1644. http://dx.doi.org/10.1093/infdis/jir611. cells after Listeria monocytogenes infection. J Immunol 178:4498 – 4505.
15. Martin-Blondel G, Bauer J, Cuvinciuc V, Uro-Coste E, Debard A, http://dx.doi.org/10.4049/jimmunol.178.7.4498.
Massip P, Delisle MB, Lassmann H, Marchou B, Mars LT, Liblau RS. 33. Lukacher AE, Wilson CS. 1998. Resistance to polyoma virus-induced
2013. In situ evidence of JC virus control by CD8⫹ T cells in PML-IRIS tumors correlates with CTL recognition of an immunodominant H-2Dk-
during HIV infection. Neurology 81:964 –970. http://dx.doi.org/10.1212 restricted epitope in the middle T protein. J Immunol 160:1724 –1734.
/WNL.0b013e3182a43e6d. 34. Moser JM, Altman JD, Lukacher AE. 2001. Antiviral CD8⫹ T cell re-
16. Gheuens S, Bord E, Kesari S, Simpson DM, Gandhi RT, Clifford DB, sponses in neonatal mice: susceptibility to polyoma virus-induced tumors
May 2016 Volume 90 Number 10 Journal of Virology jvi.asm.org 5197Qin et al.
is associated with lack of cytotoxic function by viral antigen-specific T 52. Lukacher AE, Moser JM, Hadley A, Altman JD. 1999. Visualization of
cells. J Exp Med 193:595– 606. http://dx.doi.org/10.1084/jem.193.5.595. polyoma virus-specific CD8⫹ T cells in vivo during infection and tumor
35. Kemball CC, Lee ED, Vezys V, Pearson TC, Larsen CP, Lukacher AE. rejection. J Immunol 163:3369 –3378.
2005. Late priming and variability of epitope-specific CD8⫹ T cell re- 53. Freund R, Dubensky T, Bronson R, Sotnikov A, Carroll J, Benjamin T.
sponses during a persistent virus infection. J Immunol 174:7950 –7960. 1992. Polyoma tumorigenesis in mice: evidence for dominant resistance
http://dx.doi.org/10.4049/jimmunol.174.12.7950. and dominant susceptibility genes of the host. Virology 191:724 –731.
36. Staveley-O’Carroll K, Schell TD, Jimenez M, Mylin LM, Tevethia MJ, http://dx.doi.org/10.1016/0042-6822(92)90248-N.
Schoenberger SP, Tevethia SS. 2003. In vivo ligation of CD40 enhances 54. Curtsinger JM, Lins DC, Mescher MF. 2003. Signal 3 determines toler-
priming against the endogenous tumor antigen and promotes CD8⫹ T cell ance versus full activation of naive CD8 T cells: dissociating proliferation
effector function in SV40 T antigen transgenic mice. J Immunol 171:697– and development of effector function. J Exp Med 197:1141–1151. http:
707. http://dx.doi.org/10.4049/jimmunol.171.2.697. //dx.doi.org/10.1084/jem.20021910.
37. Bender A, Sapp M, Schuler G, Steinman RM, Bhardwaj N. 1996. 55. Curtsinger JM, Valenzuela JO, Agarwal P, Lins D, Mescher MF. 2005.
Improved methods for the generation of dendritic cells from nonprolifer- Type I IFNs provide a third signal to CD8 T cells to stimulate clonal
ating progenitors in human blood. J Immunol Methods 196:121–135. expansion and differentiation. J Immunol 174:4465– 4469. http://dx.doi
http://dx.doi.org/10.1016/0022-1759(96)00079-8. .org/10.4049/jimmunol.174.8.4465.
38. Le Moine A, Flamand V, Demoor FX, Noel JC, Surquin M, Kiss R, 56. Schmidt CS, Mescher MF. 1999. Adjuvant effect of IL-12: conversion of
Nahori MA, Pretolani M, Goldman M, Abramowicz D. 1999. Critical peptide antigen administration from tolerizing to immunizing for CD8⫹
Downloaded from http://jvi.asm.org/ on January 17, 2021 by guest
roles for IL-4, IL-5, and eosinophils in chronic skin allograft rejection. J T cells in vivo. J Immunol 163:2561–2567.
Clin Invest 103:1659 –1667. http://dx.doi.org/10.1172/JCI5504. 57. Huber JP, Farrar JD. 2011. Regulation of effector and memory T-cell
39. Malmgaard L, Paludan SR, Mogensen SC, Ellermann-Eriksen S. 2000. functions by type I interferon. Immunology 132:466 – 474. http://dx.doi
Herpes simplex virus type 2 induces secretion of IL-12 by macrophages .org/10.1111/j.1365-2567.2011.03412.x.
through a mechanism involving NF-B. J Gen Virol 81:3011–3020. http: 58. Nguyen KB, Watford WT, Salomon R, Hofmann SR, Pien GC, Mori-
//dx.doi.org/10.1099/0022-1317-81-12-3011. nobu A, Gadina M, O’Shea JJ, Biron CA. 2002. Critical role for STAT4
40. Liu J, Guan X, Tamura T, Ozato K, Ma X. 2004. Synergistic activation activation by type 1 interferons in the interferon-␥ response to viral infec-
of interleukin-12 p35 gene transcription by interferon regulatory factor-1 tion. Science 297:2063–2066. http://dx.doi.org/10.1126/science.1074900.
and interferon consensus sequence-binding protein. J Biol Chem 279: 59. Luft T, Pang KC, Thomas E, Hertzog P, Hart DN, Trapani J, Cebon J.
55609 –55617. http://dx.doi.org/10.1074/jbc.M406565200. 1998. Type I IFNs enhance the terminal differentiation of dendritic cells. J
41. Hein J, Schellenberg U, Bein G, Hackstein H. 2001. Quantification of Immunol 161:1947–1953.
murine IFN-␥ mRNA and protein expression: impact of real-time kinetic 60. Marshall HD, Urban SL, Welsh RM. 2011. Virus-induced transient immune
RT-PCR using SYBR green I dye. Scand J Immunol 54:285–291. http://dx suppression and the inhibition of T cell proliferation by type I interferon. J
.doi.org/10.1046/j.1365-3083.2001.00928.x. Virol 85:5929 –5939. http://dx.doi.org/10.1128/JVI.02516-10.
42. Park SW, Kim M, Kim M, D’Agati VD, Lee HT. 2011. Sphingosine 61. Berard M, Brandt K, Bulfone-Paus S, Tough DF. 2003. IL-15 promotes
kinase 1 protects against renal ischemia-reperfusion injury in mice by the survival of naive and memory phenotype CD8⫹ T cells. J Immunol
sphingosine-1-phosphate1 receptor activation. Kidney Int 80:1315–1327. 170:5018 –5026. http://dx.doi.org/10.4049/jimmunol.170.10.5018.
http://dx.doi.org/10.1038/ki.2011.281. 62. Mattei F, Schiavoni G, Belardelli F, Tough DF. 2001. IL-15 is expressed
43. Soni C, Domeier PP, Wong EB, Shwetank, Khan TN, Elias MJ, Schell by dendritic cells in response to type I IFN, double-stranded RNA, or
SL, Lukacher AE, Cooper TK, Rahman ZS. 2015. Distinct and synergistic lipopolysaccharide and promotes dendritic cell activation. J Immunol
roles of Fc␥RIIB deficiency and 129 strain-derived SLAM family proteins 167:1179 –1187. http://dx.doi.org/10.4049/jimmunol.167.3.1179.
in the development of spontaneous germinal centers and autoimmunity. J 63. Zhang X, Sun S, Hwang I, Tough DF, Sprent J. 1998. Potent and
Autoimmun 63:31– 46. http://dx.doi.org/10.1016/j.jaut.2015.06.011. selective stimulation of memory-phenotype CD8⫹ T cells in vivo by IL-15.
44. Livak KJ, Schmittgen TD. 2001. Analysis of relative gene expression data Immunity 8:591–599. http://dx.doi.org/10.1016/S1074-7613(00)80564-6.
using real-time quantitative PCR and the 2⫺⌬⌬Ct method. Methods 25: 64. Martinet V, Tonon S, Torres D, Azouz A, Nguyen M, Kohler A,
402– 408. http://dx.doi.org/10.1006/meth.2001.1262. Flamand V, Mao CA, Klein WH, Leo O, Goriely S. 2015. Type I
45. Pallas DC, Schley C, Mahoney M, Harlow E, Schaffhausen BS, Roberts interferons regulate eomesodermin expression and the development of
TM. 1986. Polyomavirus small t antigen: overproduction in bacteria, pu- unconventional memory CD8⫹ T cells. Nat Commun 6:7089. http://dx
rification, and utilization for monoclonal and polyclonal antibody pro- .doi.org/10.1038/ncomms8089.
duction. J Virol 60:1075–1084. 65. Honke N, Shaabani N, Merches K, Gassa A, Kraft A, Ehrhardt K,
46. Akondy RS, Johnson PL, Nakaya HI, Edupuganti S, Mulligan MJ, Haussinger D, Lohning M, Dittmer U, Hengel H, Recher M, Lang PA,
Lawson B, Miller JD, Pulendran B, Antia R, Ahmed R. 2015. Initial viral Lang KS. 2016. Immunoactivation induced by chronic viral infection
load determines the magnitude of the human CD8 T cell response to inhibits viral replication and drives immunosuppression through sus-
yellow fever vaccination. Proc Natl Acad Sci U S A 112:3050 –3055. http: tained IFN-I responses. Eur J Immunol 46:372–380. http://dx.doi.org/10
//dx.doi.org/10.1073/pnas.1500475112. .1002/eji.201545765.
47. Wherry EJ, Blattman JN, Ahmed R. 2005. Low CD8 T-cell proliferative 66. Odorizzi PM, Wherry EJ. 2013. Immunology. An interferon paradox.
potential and high viral load limit the effectiveness of therapeutic vaccina- Science 340:155–156. http://dx.doi.org/10.1126/science.1237568.
tion. J Virol 79:8960 – 8968. http://dx.doi.org/10.1128/JVI.79.14.8960 67. Teijaro JR, Ng C, Lee AM, Sullivan BM, Sheehan KC, Welch M,
-8968.2005. Schreiber RD, de la Torre JC, Oldstone MB. 2013. Persistent LCMV
48. Kemball CC, Pack CD, Guay HM, Li ZN, Steinhauer DA, Szomolanyi- infection is controlled by blockade of type I interferon signaling. Science
Tsuda E, Lukacher AE. 2007. The antiviral CD8⫹ T cell response is 340:207–211. http://dx.doi.org/10.1126/science.1235214.
differentially dependent on CD4⫹ T cell help over the course of persistent 68. Wiesel M, Crouse J, Bedenikovic G, Sutherland A, Joller N, Oxenius A.
infection. J Immunol 179:1113–1121. http://dx.doi.org/10.4049 2012. Type-I IFN drives the differentiation of short-lived effector CD8⫹ T
/jimmunol.179.2.1113. cells in vivo. Eur J Immunol 42:320 –329. http://dx.doi.org/10.1002/eji
49. Schluns KS, Kieper WC, Jameson SC, Lefrancois L. 2000. Interleukin-7 .201142091.
mediates the homeostasis of naive and memory CD8 T cells in vivo. Nat 69. Lukacher AE, Ma Y, Carroll JP, Abromson-Leeman SR, Laning JC, Dorf
Immunol 1:426 – 432. http://dx.doi.org/10.1038/80868. ME, Benjamin TL. 1995. Susceptibility to tumors induced by polyoma virus
50. Kaech SM, Tan JT, Wherry EJ, Konieczny BT, Surh CD, Ahmed R. is conferred by an endogenous mouse mammary tumor virus superantigen. J
2003. Selective expression of the interleukin 7 receptor identifies effector Exp Med 181:1683–1692. http://dx.doi.org/10.1084/jem.181.5.1683.
CD8 T cells that give rise to long-lived memory cells. Nat Immunol 70. Stroh LJ, Maginnis MS, Blaum BS, Nelson CD, Neu U, Gee GV, O’Hara
4:1191–1198. http://dx.doi.org/10.1038/ni1009. BA, Motamedi N, DiMaio D, Atwood WJ, Stehle T. 2015. The greater
51. Joshi NS, Cui W, Chandele A, Lee HK, Urso DR, Hagman J, Gapin L, affinity of JC polyomavirus capsid for ␣2,6-linked lactoseries tetrasaccha-
Kaech SM. 2007. Inflammation directs memory precursor and short-lived ride c than for other sialylated glycans is a major determinant of infectivity.
effector CD8⫹ T cell fates via the graded expression of T-bet transcription J Virol 89:6364 – 6375. http://dx.doi.org/10.1128/JVI.00489-15.
factor. Immunity 27:281–295. http://dx.doi.org/10.1016/j.immuni.2007.07 71. Ray U, Cinque P, Gerevini S, Longo V, Lazzarin A, Schippling S, Martin
.010. R, Buck CB, Pastrana DV. 2015. JC polyomavirus mutants escape anti-
5198 jvi.asm.org Journal of Virology May 2016 Volume 90 Number 10You can also read
